Sie sind auf Seite 1von 6

The Journal of Nutrition

Nutritional Epidemiology

Dietary Fat Intake Modifies the Effect of a


Common Variant in the LIPC Gene on Changes
in Serum Lipid Concentrations during a
Long-Term Weight-Loss Intervention Trial1–3
Min Xu,4,5* San San Ng,7 George A Bray,8 Donna H Ryan,8 Frank M Sacks,4 Guang Ning,5,6 and Lu Qi4,9,10*

Downloaded from https://academic.oup.com/jn/article-abstract/145/6/1289/4616777 by guest on 13 June 2020


4
Department of Nutrition, Harvard School of Public Health, Boston, MA; 5Shanghai Clinical Center for Endocrine and Metabolic Diseases,
Shanghai Institute of Endocrine and Metabolic Diseases, Shanghai, China; 6Department of Endocrinology and Metabolism, Ruijin Hospital
Affiliated with Shanghai Jiao Tong University School of Medicine, Shanghai, China; 7Department of Molecular Genetics, GenoVive, New
Orleans, LA; 8Pennington Biomedical Research Center, Louisiana State University System, Baton Rouge, LA; 9Channing Division of Network
Medicine, Department of Medicine, Brigham and WomenÕs Hospital, Boston, MA; and 10Harvard Medical School, Boston, MA

Abstract
Background: Hepatic lipase (HL) plays a pivotal role in the metabolism of HDL and LDL. Recent genome-wide association
studies have identified common variants in the HL gene (LIPC) associated with HDL cholesterol.
Objective: We tested the effect of a common variant in LIPC on changes in blood lipids in response to weight-loss diets in
the Preventing Overweight Using Novel Dietary Strategies Trial.
Methods: We genotyped LIPC rs2070895 in 743 overweight or obese adults aged 30–70 y (61% women) who were
assigned to high-fat (40% energy) or low-fat (20% energy) diets for 2 y. We measured serum concentrations of total
cholesterol (TC), triglycerides, LDL cholesterol, and HDL cholesterol at baseline and 2 y of intervention.
Results: At 2 y of intervention, dietary fat modified effects of the variant on changes in serum TC, LDL cholesterol, and
HDL cholesterol (P-interaction: 0.0008, 0.004, and 0.03, respectively). In the low-fat group, as compared to the G allele, the
A allele tended to be related to the decrease in TC and LDL cholesterol concentrations [TC (b 6 SE): 25.5 6 3.0, P = 0.07;
LDL cholesterol: 24.8 6 2.5, P = 0.06] and a lower increase in HDL cholesterol concentrations (b 6 SE: 21.37 6 0.69, P =
0.048), whereas an opposite effect in the high-fat diet group was evident [TC (b 6 SE): 7.3 6 2.7, P = 0.008; LDL
cholesterol: 4.1 6 2.3, P = 0.07], and there was no genetic effect on changes in HDL cholesterol concentrations (P = 0.54).
Conclusion: Dietary fat intake modifies the effect of a common variant in LIPC on changes in serum lipids during a
long-term weight-loss intervention in overweight or obese adults. This trial was registered at clinicaltrials.gov as
NCT00072995. J Nutr 2015;145:1289–94.

Keywords: hepatic lipase, LIPC, high-fat diet, gene-diet interaction, weight-loss intervention

Introduction
cardiovascular risk. Hepatic lipase (HL)11, which is expressed in
Dyslipidemia, such as low HDL cholesterol, and high LDL the liver, plays a pivotal role in the metabolism of lipoproteins as
cholesterol and TGs, has been well associated with increased a lipolytic enzyme that hydrolyzes TGs and phospholipids in
chylomicron remnants, intermediate-density lipoprotein, and
1
Supported by grants from the National Heart, Lung, and Blood Institute
HDL. HL has a multifunctional effect on the development of
(HL071981), the National Institute of Diabetes and Digestive and Kidney Diseases atherosclerosis. Previous studies support proatherogenic and
(DK091718), the General Clinical Research Center (RR-02635), the Boston Obesity antiatherogenic functions for HL (1).Candidate gene studies
Nutrition Research Center (DK46200), the United States–Israel Binational Science have reported that variation in the HL gene (LIPC) affected the
Foundation (2011036), the National Natural Science Foundation of China (81270877
effectiveness of the therapy in reducing the risk of cardiovascular
and 81471062), and the Shanghai Pujiang Project (14PJD024). L Qi was a recipient
of the AHA Scientist Development Award (0730094N). disease (2, 3). Recent genome-wide association studies have
2
Author disclosures: M Xu, SS Ng, GA Bray, DH Ryan, FM Sacks, G Ning, and identified and confirmed a common variant rs2070895 in LIPC
L Qi, no conflicts of interest.
3
Supplemental Tables 1 and 2 are available from the ‘‘Online Supporting
Material’’ link in the online posting of the article and from the same link in the
online table of contents at http://jn.nutrition.org. 11
Abbreviations used: HL, hepatic lipase; LIPC, hepatic lipase gene; POUNDS
* To whom correspondence should be addressed. E-mail: nhlqi@channing. LOST, Preventing Overweight Using Novel Dietary Strategies; SNP, single
harvard.edu (L Qi), della.xumin@163.com (M Xu). nucleotide polymorphism; TC, total cholesterol.

ã 2015 American Society for Nutrition.


Manuscript received February 23, 2015. Initial review completed March 5, 2015. Revision accepted April 3, 2015. 1289
First published online April 29, 2015; doi:10.3945/jn.115.212514.
that is associated with HDL cholesterol concentrations (4, 5), clinical laboratory at the Pennington Biomedical Research Center. TG,
probably through affecting the activity of HL. A family study TC, LDL cholesterol, and HDL cholesterol were measured on the
showed that the total allelic variations at LIPC might account Synchron CX7 (Beckman Coulter). LDL was calculated for each
for up to 25% of the total variance in plasma HDL cholesterol participant according to the following formula: LDL cholesterol = TC
2 HDL cholesterol 2 TG / 5 (14), except when TG concentration
concentrations (6).
exceeded 400 mg/dL, in which case LDL cholesterol was measured
Several previous studies found significant interaction between directly on all samples of the participant. BMI was calculated as weight in
LIPC genetic variants and dietary fat intake on the blood kilograms divided by height in squared meters.
concentrations of lipids such as HDL cholesterol (7–11). How-
ever, those studies were largely observational in nature and Genotyping. DNA was extracted from the buffy coat fraction of
might be biased by reverse causation and confounding. To our centrifuged blood by using the QIAmp Blood Kit (Qiagen). We genotyped
knowledge, no study has assessed the long-term effect of LIPC single nucleotide polymorphism (SNP) rs2070895 near LIPC that was
variants on changes in lipid profile and the potential interactions associated with the serum lipid concentrations in recent genome-wide
with dietary fat intake in randomized clinical trials. association studies (4). Genotyping was performed with use of the
In the present study, we tested the effects of the common OpenArray SNP Genotyping System (BioTrove), and the genotyping
success rate was 99%. Replicated quality control samples (10%) were
variant rs2070895 in LIPC on changes in blood lipid concen-
included in every genotyping plate with >99% concordance (15).

Downloaded from https://academic.oup.com/jn/article-abstract/145/6/1289/4616777 by guest on 13 June 2020


trations in response to a 2-y diet intervention in a randomized
clinical trial, the Preventing Overweight Using Novel Dietary Statistical analysis. We used SAS version 9.1 (SAS Institute, Inc.) to
Strategies (POUNDS LOST) Trial, and particularly examined perform the data analysis. Because we aimed to examine the long-term
the gene-diet interactions. changes in blood lipids during the intervention, the primary outcomes of
the analysis were 2-y changes in blood lipids. Baseline data are presented
as means 6 SDs or medians [IQRs] for the continuous variables and n
Methods (%) for the categorical variables. Variables with skewed distribution
Participants. The POUNDS LOST Trial (NCT00072995) was designed (serum TG) were log transformed before analysis. The Hardy-Weinberg
to compare the effects of energy-reduced diets with different composi- equilibrium of rs2070895 genotypes and comparison of categorical
tions of macronutrients on body weight during 2 y of follow-up. The variables at baseline were accessed by x2 test. We used ANCOVA to test
study design has been described in detail elsewhere (12, 13). Briefly, 811 the differences in continuous variables at baseline, with adjustment for
overweight and obese subjects (25 # BMI < 40 kg/m2) aged 30–70 y with age, sex, and ethnicity groups. The primary outcomes were changes
no diabetes or unstable cardiovascular disease, no use of medications (i.e., the concentrations at 2 y of follow-up time minus the baseline
affecting body weight, and insufficient motivation to take part in the concentrations) in fasting serum TC, TG, HDL cholesterol, and LDL
trial, were randomly assigned to 4 diets; the target percentages of energy cholesterol over the time the participant remained in the trial. Multi-
derived from fat, protein, and carbohydrate in the 4 diets were respectively variate general linear models were used to test the main genetic effects
20%, 15%, and 65%; 20%, 25%, and 55%; 40%, 15%, and 45%; and and the potential interactions between the genetic variation and diet
40%, 25%, and 35%. Thus, 2 diets were low fat (20%) and 2 diets were intervention (high vs. low fat) on changes in weight and lipid profile (TC,
high fat (40%), or 2 diets were average protein (15%) and 2 diets were TG, HDL cholesterol, and LDL cholesterol). The significance of gene-diet
high protein (25%), which constituted a 2-by-2 factorial design. The intervention interactions was tested by including the genotype-by-diet
nutrient goals of the 4 diets were to be low in saturated fat and cholesterol interaction multiplicative terms in the models. The covariates included age,
and high in fiber. In detail, the diets included 8% saturated fat for each sex, ethnicity, baseline BMI, and weight loss, and values for the respective
group; 6% monounsaturated fat for low-fat and 22% for high-fat groups; outcome were used in the analysis. Linear mixed models were used to test
6% polyunsaturated fat for low-fat and 10% for high-fat groups; at least genetic associations with the trajectory of changes in weight or lipids
20 g dietary fiber/d; and #150 mg cholesterol/1000 kcal. Carbohydrate- according to diet groups. Time was treated as a repeated measurement
rich foods with a lower glycemic index were used. Each participantÕs diet factor, with which genotype-time interaction terms were included in the
prescription represented a 750-kcal/d deficit from baseline, which was mixed models. An additive genetic model was analyzed. Because the
calculated from the personÕs resting energy expenditure and activity level. majority of study participants were white (80%), we also performed
After 2 y, 80% of the participants (n = 645) completed the trial. The study sensitivity analysis in white participants. All P values were two-sided. A P
was approved by the human subjects committee at the Harvard School of value of 0.05 was considered statistically significant.
Public Health and Brigham and WomenÕs Hospital and the Pennington
Biomedical Research Center of the Louisiana State University System and
by a data and safety monitoring board appointed by the National Heart,
Lung, and Blood Institute. All participants gave written informed consent. Results
The present study reported the secondary outcomes of the POUNDS
The mean age of the total participants was 51 y and mean
LOST Trial. We included 743 participantsÕ genotype data available at baseline
BMI was 32.7 kg/m2. Among the participants, 61% were
in the current study. There was no significant difference in the baseline
characteristics between the participants with and without genotype data.
women, 80% were white, 15% were black, 3% were Hispanic,
and 2% were Asian or other ethnic groups. The minor allele
Assessment of outcomes and covariates. Body height was measured (A) frequency of LIPC rs2070895 was 26.4% in total partic-
at baseline. Body weight and waist circumference were measured in the ipants, and the genotype distribution fits the Hardy-Weinberg
morning before breakfast at each intervention visit. Blood pressure was equilibrium.
measured on 2 d, at baseline, 6 mo, and 2 y, by using an automated Table 1 shows baseline characteristics of participants
device (HEM-907XL; Omron). Dietary intake was assessed in a random according to the LIPC rs2070895 genotypes. Genotype fre-
sample of 50% of the participants by a review of 5-d diet record at quencies were similar between men and women and across diet
baseline and by 24-h recall during a telephone interview on 3 noncon-
groups. Significant difference of the genotype distribution was
secutive days at 6 mo and 2 y. Respiratory quotient was measured at
observed among the ethnic groups (P < 0.0001). The A allele
baseline and during follow-up at 6 mo and 2 y as a biomarker of
adherence to the assigned diet, i.e., specific carbohydrate, protein, and fat
of SNP rs2070895 was related to higher concentrations of
content. Fasting blood samples, 24-h urine samples, and measurement of HDL cholesterol. No other differences in baseline characteristics
resting metabolic rate were obtained on 1 d. Concentrations of fasting across the genotype were observed (all P $ 0.05). The results
serum glucose, TG, total cholesterol (TC), HDL cholesterol, and LDL were similar in the white participants (Supplemental Table 1). In
cholesterol and 24-h urinary nitrogen excretion were measured at the addition, there were no significant differences in nutrient intakes
1290 Xu et al.
TABLE 1 Baseline characteristics of the participants according to LIPC rs2070895 genotypes1

AA (n = 62) AG (n = 268) GG (n = 413) P2

Age, y 47.9 6 8.3 50.5 6 9.1 51.9 6 9.4 0.06


Female 21 (33.9) 103 (38.4) 166 (40.2) 0.62
Dietary fat group 0.83
Low fat 30 (8.1) 138 (37.1) 204 (54.8)
High fat 32 (8.6) 130 (35.0) 209 (56.3)
Race or ethnic group ,0.0001
White 31 (5.2) 198 (33.3) 366 (61.5)
Black 27 (24.1) 53 (47.3) 32 (28.6)
Hispanic 3 (12.0) 13 (52.0) 9 (36.0)
Asian or other 1 (9.0) 4 (36.4) 6 (54.6)
Height, cm 167 6 8.5 169 6 9.2 169 6 8.5 0.09
Weight, kg 91.8 6 14.2 93.5 6 15.2 93.3 6 16.0 0.22

Downloaded from https://academic.oup.com/jn/article-abstract/145/6/1289/4616777 by guest on 13 June 2020


Waist circumference, cm 103 6 12.5 103 6 12.9 104 6 13.3 0.75
BMI, kg/m2 33.0 6 3.8 32.7 6 3.8 32.6 6 3.9 0.75
Blood pressure, mm Hg
Systolic 122 6 16 120 6 13 119 6 13 0.05
Diastolic 77 6 11 76 6 9 75 6 9 0.20
Fasting serum glucose, mg/dL 93 6 16 91 6 11 92 6 12 0.88
Fasting serum lipid profile, mg/dL
TGs 98.5 [70.0–168] 121 [80.0–165] 124 [90.0–193] 0.17
TC 205 6 41.3 201 6 37.1 203 6 36.1 0.47
HDL cholesterol 51.0 6 14.3 49.6 6 15.6 47.9 6 13.0 0.002
LDL cholesterol 129 6 34.2 125 6 31.3 126 6 32.2 0.99
Dietary intake per day
Carbohydrate, % of energy 45.9 6 7.5 45.2 6 8.0 44.1 6 7.4 0.28
Fat, % of energy 35.8 6 6.9 37.0 6 6.3 37.1 6 5.7 0.48
Protein, % of energy 18.3 6 3.2 17.7 6 3.3 18.4 6 3.4 0.17
Energy, kcal 1881 6 572 2036 6 539 1936 6 569 0.33
Urinary nitrogen, g/d 11.3 6 4.6 12.2 6 4.5 12.4 6 4.3 0.54
Respiratory quotient 0.84 6 0.05 0.84 6 0.04 0.84 6 0.04 0.72
1
Values are means 6 SDs, medians [IQRs], or n (%). LIPC, hepatic lipase gene; TC, total cholesterol.
2
P values were calculated by x2 test for categorical variables and ANCOVA for continuous variables after adjusting for age, sex, and
ethnicity.

and biomarkers of adherence to the assigned diet at 2 y across genetic variation and lipids. We did not find a significant
the LIPC rs2070895 genotype in low- and high-fat diet groups. difference in changes of TC after further adjustment for changes
The weight losses (means 6 SDs) in the high- and low-fat in LDL cholesterol at 2 y in the low-fat or high-fat groups. We
diet groups were 4.0 6 7.3 kg and 4.2 6 7.6 kg at 2 y, found opposite genetic effects on changes in HDL cholesterol
respectively. The corresponding weight losses among the 3 concentrations. The A allele was significantly associated with
genotypes of LIPC rs2070895 variant were 2.9 6 4.9 kg in the less increase of HDL cholesterol concentrations in the low-fat
AA, 3.8 6 7.6 kg in the GA, and 4.4 6 7.6 kg in the GG group (b 6 SE: 21.37 6 0.69, P = 0.048), whereas no
genotype at 2 y, respectively. No main genetic effect or gene- association was observed in the high-fat group (b 6 SE: 0.44 6
diet interaction was found on weight loss during 2 y of 0.71, P = 0.54). We did not detect any interactions between the
intervention. Genotype did not affect the changes in serum TC, genetic variation and dietary fat intake on changes in serum TG
LDL cholesterol, or HDL cholesterol concentrations during 2 y at 2 y of intervention (P-interaction: 0.62). Similar interactions
of intervention (all P $ 0.35). were observed when the analysis was restricted to the white
We then examined the interaction of dietary fat intake and participants (Supplemental Table 2).
the genetic variation on changes in serum lipids. At 2 y of We then examined the trajectory changes in lipids accord-
intervention, we observed that dietary fat significantly modified ing to the LIPC rs2070895 genotypes by dietary fat intake
the genetic effects of rs2070895 on changes in serum TC, LDL over the 2-y intervention period. We observed significant
cholesterol, and HDL cholesterol (P-interaction: 0.0008, 0.004, genotype-time interactions on changes in TC, LDL cholesterol,
and 0.03, respectively), with adjustment for age, sex, ethnicity, and HDL cholesterol in the low-fat diet group (P-interactions:
baseline BMI, weight loss, and baseline measurements for the 0.01, 0.01, and 0.005, respectively). The LIPC genetic effects
respective outcomes (Figure 1, Table 2). In the low-fat diet on TC and LDL cholesterol displayed a cumulative pattern
group, the A allele was associated with a decrease of TC and throughout the intervention and reached the maximum values
LDL cholesterol concentrations [TC (b 6 SE): 25.5 6 3.0, P = at 2 y (all P-trend values were #0.002 for AA and AG
0.07; LDL cholesterol: 24.8 6 2.5, P = 0.06], whereas an genotype, but there were no significant trends for GG
opposite genetic effect was found in the high-fat diet group [TC genotype), whereas continued cumulative increasing trends
(b 6 SE): 7.3 6 2.7, P = 0.008; LDL cholesterol: 4.1 6 2.3, P = were found in HDL cholesterol concentrations (both P-trend
0.07]. Then we tested the independent associations between the values were #0.0001 for GG and AG genotype, but there were
Diet modifies LIPC gene effect on serum lipids 1291
FIGURE 1 Trajectory analysis on 2-y
changes in serum total cholesterol (A, D),
LDL cholesterol (B, E), and HDL cholesterol
(C, F) by LIPC rs2070895 genotype of
participants that consumed the low-fat (up-
per panels) or high-fat (lower panels) diets.
The number of participants in the low-fat diet
group (A–C), for AA: n = 30 at baseline, n =
22 at 6 mo, and n = 20 at 2 y; AG: n = 138 at
baseline, n = 118 at 6 mo, and n = 101 at 2 y;
and GG: n = 204 at baseline, n = 182 at 6
mo, and n = 155 at 2 y. The corresponding
numbers in the high-fat diet group (D–F), AA:
n = 32, 24, and 20, at baseline, 6 mo, and
2 y, respectively; AG: n = 130, 110, and 98,
at baseline, 6 mo, and 2 y, respectively; and
GG: n = 209,180, and 149, at baseline, 6 mo,

Downloaded from https://academic.oup.com/jn/article-abstract/145/6/1289/4616777 by guest on 13 June 2020


and 2 y, respectively. Values are adjusted
means 6 SEs after adjusting for age, sex, ethnicity, baseline BMI, weight loss, and baseline values for respective outcomes. P-interactions were
derived from the linear mixed models, in which intervention time was treated as a repeated measurement factor, and genotype time as the
interaction terms. LIPC, hepatic lipase gene.

no significant trends for GG genotype) (Figure 1A–C). In the allele. In the current study, the minor allele A was associated
high-fat diet group, we observed a significant genotype-time with less of an increase in HDL cholesterol in response to a low-
interaction on changes in TC (P-interaction: 0.03), a marginal fat diet. Our data suggest that LIPC may partly account for the
interaction on changes in LDL cholesterol (P-interaction: 0.06), interindividual heterogeneity in HDL cholesterol changes in-
and no significant interaction on changes in HDL cholesterol (P- duced by different dietary fat intake, although the underlying
interaction: 0.68) (Figure 1D–F). mechanisms remain to be determined.
In addition to HDL cholesterol, we also found that dietary fat
intake significantly modified the genetic effect of LIPC on
Discussion changes in TC and LDL cholesterol concentrations during the
In this 2-y randomized weight-loss intervention trial, we weight-loss trial. Our results were consistent with previous
observed a significant interaction between the LIPC SNP investigation (19). Lindi et al. (19) reported that individuals with
rs2070895 and dietary fat intake on changes in serum TC, the AA genotype were responsive to the multiunsaturated FA–
LDL cholesterol, and HDL cholesterol concentrations. The A enriched diet, and their serum LDL cholesterol concentrations
allele was related to a decrease in TC and LDL cholesterol and a decreased more than in subjects with other genotypes. Interest-
small increase in HDL cholesterol in response to the low-fat diet, ingly, we found that the genetic effect on changes in TC and LDL
whereas an opposite genetic effect was observed when partic- cholesterol in response to a low-fat diet showed a cumulative
ipants were assigned a high-fat diet. The genetic effect on pattern throughout the intervention and reached the maximum
changes of blood cholesterol concentrations showed a long-
term, cumulative pattern throughout the intervention. TABLE 2 Dietary fat intakes modulated the genotype effect of
Our findings are in line with several previous studies. These the LIPC rs2070895 variant on the changes in lipids in response to
studies have reported that the effects of 2514C/T (rs1800588) the dietary fat intervention at 2 y1
of LIPC, which is in high linkage disequilibrium with rs2070895
Low-fat diet High-fat diet P-interaction
(r2 = 0.97), on HDL cholesterol concentrations were modified by
dietary fat intake (7, 10, 11). Like rs2070895, rs1800588 is DTotal cholesterol, mg/dL
2

located in the promoter region and associated with decreased AA 20 (224.7 6 8.2) 20 (7.4 6 9.5)
plasma HL activity and increased HDL cholesterol concentra- AG 101 (215.0 6 5.2) 98 (1.3 6 7.8)
tions (3, 16, 17). In the Framingham Study, the T allele (in GG 155 (211.2 6 5.2) 149 (26.3 6 7.6)
linkage disequilibrium with allele A in rs2070895) was associ- P-trend3 0.07 0.008 0.0008
ated with significantly higher HDL cholesterol concentrations DLDL cholesterol,2 mg/dL
and large particle size only in subjects consuming 30% of energy AA 20 (221.8 6 6.8) 20 (10.7 6 8.1)
from fat (10). The study indicated an intraindividual difference AG 101 (211.2 6 4.3) 98 (6.3 6 6.7)
in the plasma lipid response to the dietary fat. Similar results GG 155 (28.7 6 4.3) 149 (2.1 6 6.5)
were found in African Americans (16) and subjects of Indian P-trend3 0.06 0.07 0.004
origin living in Singapore (11). DHDL cholesterol,2 mg/dL
However, few studies have assessed interactions between AA 20 (2.73 6 1.85) 20 (5.67 6 2.42)
LIPC genotype and dietary factors in intervention trials (18). In AG 101 (3.32 6 1.16) 98 (9.02 6 1.99)
the Finnish Diabetes Prevention Study (18), the A allele was GG 155 (4.81 6 1.16) 149 (7.09 6 1.95)
associated with a low conversion rate to type 2 diabetes in the P-trend3 0.048 0.54 0.03
lifestyle intervention group, which was assigned to a reduced fat
intake of <30% of energy consumed and reduced intake of
1
Values are n (means 6 SEs). LIPC, hepatic lipase gene.
2
The D variables indicate the change of respective outcomes.
saturated fat to <10% of energy consumed. In addition, a greater 3
P-trend was for general linear regression in which an additive genetic model was
increase in HDL cholesterol concentration from baseline to 3 y analyzed for the rs2070895 variant. The adjustment was including age, sex, ethnicity,
was found in subjects carrying the A allele than those with the G baseline BMI, weight loss, and baseline values for respective outcomes.

1292 Xu et al.
values at 2 y. The gene-time interactions were less significant in 2. Peters BJ, Pett H, Klungel OH, Stricker BH, Psaty BM, Glazer NL,
the high-fat diet. These findings suggest that the gene-diet Wiggins KL, Bis JC, de Boer A, Maitland-van der Zee AH. Genetic
variability within the cholesterol lowering pathway and the effective-
interactions on the changes of cholesterol concentrations might ness of statins in reducing the risk of MI. Atherosclerosis 2011;217:
be persistent. 458–64.
HL plays a pivotal role in lipid metabolism by affecting both 3. Zambon A, Deeb SS, Pauletto P, Crepaldi G, Brunzell JD. Hepatic
reverse cholesterol transport and remodeling of TG-rich lipo- lipase: a marker for cardiovascular disease risk and response to therapy.
protein particles and, thereby, the formation of atherogenic small Curr Opin Lipidol 2003;14:179–89.
dense LDLs (1). Increased HL is associated with smaller and 4. Teslovich TM, Musunuru K, Smith AV, Edmondson AC, Stylianou IM,
denser LDL and HDL particles that are proatherogenic, whereas Koseki M, Pirruccello JP, Ripatti S, Chasman DI, Willer CJ, et al.
Biological, clinical and population relevance of 95 loci for blood lipids.
decreased HL is associated with larger and more buoyant LDL Nature 2010;466:707–13.
and HDL particles (20). Previous animal studies showed that the 5. Kraja AT, Vaidya D, Pankow JS, Goodarzi MO, Assimes TL, Kullo IJ,
activity of HL in the liver could be depressed by high-fat feeding Sovio U, Mathias RA, Sun YV, Franceschini N, et al. A bivariate
(21). Dysfunctional HL may associate with elevated blood HDL genome-wide approach to metabolic syndrome: STAMPEED Consor-
cholesterol concentrations with varying size of HDL particles tium. Diabetes 2011;60:1329–39.
and paradoxically increased atherosclerosis risk. In the present 6. Cohen JC, Wang Z, Grundy SM, Stoesz MR, Guerra R. Variation at the
hepatic lipase and apolipoprotein AI/CIII/AIV loci is a major cause of

Downloaded from https://academic.oup.com/jn/article-abstract/145/6/1289/4616777 by guest on 13 June 2020


study, we observed a greater decrease in TC and LDL cholesterol
genetically determined variation in plasma HDL cholesterol levels.
and a small increase in HDL cholesterol in the A allele carriers J Clin Invest 1994;94:2377–84.
assigned to the low-fat diet, which may give suggestive implica- 7. Zhang C, Lopez-Ridaura R, Rimm EB, Rifai N, Hunter DJ, Hu FB.
tions in preventive medicine and public health. Interactions between the 2514C/T polymorphism of the hepatic
To our knowledge, this is the first report about the interac- lipase gene and lifestyle factors in relation to HDL concentrations
tions of the genetic effect of the HL activity determinant gene among US diabetic men. Am J Clin Nutr 2005;81:1429–35.
and dietary fat intake on changes of serum lipids, especially 8. Bos G, Dekker JM, Feskens EJ, Ocke MC, Nijpels G, Stehouwer CD,
Bouter LM, Heine RJ, Jansen H. Interactions of dietary fat intake and
cholesterols, in a long-term, randomized weight-loss interven-
the hepatic lipase –480C/T polymorphism in determining hepatic
tion trial. Our results indicate that the effect of genetic variation lipase activity: the Hoorn Study. Am J Clin Nutr 2005;81:911–5.
of LIPC on changes of lipid concentrations was modified by 9. Deeb SS, Zambon A, Carr MC, Ayyobi AF, Brunzell JD. Hepatic lipase
dietary fat intake. In contrast to observational studies, the study and dyslipidemia: interactions among genetic variants, obesity, gender,
conditions in randomized clinical trials are controlled directly and diet. J Lipid Res 2003;44:1279–86.
by the investigators, including the specifically defined dietary 10. Ordovas JM, Corella D, Demissie S, Cupples LA, Couture P, Coltell O,
intakes. This control minimizes the possibility of bias and increases Wilson PW, Schaefer EJ, Tucker KL. Dietary fat intake determines the
effect of a common polymorphism in the hepatic lipase gene promoter
the level of causality (22). on high-density lipoprotein metabolism: evidence of a strong dose effect
However, some limitations should be acknowledged. We did in this gene-nutrient interaction in the Framingham Study. Circulation
not measure the HL activity, which did not allow assessing the 2002;106:2315–21.
direct genetic effect of LIPC on HL. However, a genetic marker 11. Tai ES, Corella D, Deurenberg-Yap M, Cutter J, Chew SK, Tan CE,
could be a surrogate for the biomarker in yielding causal relation Ordovas JM; Singapore National Health Survey. Dietary fat interacts
according to the Mendelian randomization principle (23, 24). with the -514C>T polymorphism in the hepatic lipase gene promoter on
plasma lipid profiles in a multiethnic Asian population: the 1998
Moreover, previous studies have shown that the minor alleles of Singapore National Health Survey. J Nutr 2003;133:3399–408.
these promoter variants (rs2070895 and rs1800588) have been 12. Sacks FM, Bray GA, Carey VJ, Smith SR, Ryan DH, Anton SD,
associated with diminished transcriptional activity in vitro (25) McManus K, Champagne CM, Bishop LM, Laranjo N, et al. Compar-
and a 15–45% decrease in plasma HL activity in vivo (26). Also, ison of weight-loss diets with different compositions of fat, protein, and
because the participants are all overweight or obese, our results carbohydrates. N Engl J Med 2009;360:859–73.
should be cautiously generalized to the general population with 13. Qi Q, Bray GA, Smith SR, Hu FB, Sacks FM, Qi L. Insulin receptor
substrate 1 gene variation modifies insulin resistance response to
normal body weight. Finally, given that 80% of the participants
weight-loss diets in a 2-year randomized trial: the Preventing Over-
are white, future studies are warranted to verify the gene-diet weight Using Novel Dietary Strategies (POUNDS LOST) Trial. Circu-
interactions in other ethnicity groups. lation 2011;124:563–71.
In conclusion, we observed that dietary fat intake signif- 14. Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concen-
icantly modified the LIPC rs2070895 genetic effect on changes tration of low-density lipoprotein cholesterol in plasma, without use of
in serum TC, LDL cholesterol, and HDL cholesterol concentra- the preparative ultracentrifuge. Clin Chem 1972;18:499–502.
tions during a 2-y randomized weight-loss intervention trial. 15. Xu M, Qi Q, Liang J, Bray GA, Hu FB, Sacks FM, Qi L. Genetic
determinant for amino acid metabolites and changes in body weight and
Our data suggest that the carriers of A alleles might benefit more
insulin resistance in response to weight-loss diets: the Preventing
in terms of improved lipid profiles by eating a low-fat (20%), Overweight Using Novel Dietary Strategies (POUNDS LOST) Trial.
high-carbohydrate (55–65%) diet. Circulation 2013;127:1283–9.
16. Nettleton JA, Steffen LM, Ballantyne CM, Boerwinkle E, Folsom AR.
Acknowledgments Associations between HDL-cholesterol and polymorphisms in hepatic
MX and LQ designed the research; MX and SSN conducted the lipase and lipoprotein lipase genes are modified by dietary fat intake in
African American and White adults. Atherosclerosis 2007;194:e131–40.
research and analyzed the data; GAB, DHR, FMS, and GN
17. Isaacs A, Sayed-Tabatabaei FA, Njajou OT, Witteman JC, van Duijn
provided essential materials; MX and LQ wrote the paper; and CM. The -514 C->T hepatic lipase promoter region polymorphism and
LQ had primary responsibility for final content. All authors plasma lipids: a meta-analysis. J Clin Endocrinol Metab 2004;89:
read and approved the final manuscript. 3858–63.
18. Todorova B, Kubaszek A, Pihlajamäki J, Lindström J, Eriksson J, Valle
TT, Hämäläinen H, Ilanne-Parikka P, Keinänen-Kiukaanniemi S,
References Tuomilehto J, et al.; Finnish Diabetes Prevention Study. The G-250A
promoter polymorphism of the hepatic lipase gene predicts the
1. Santamarina-Fojo S, Haudenschild C, Amar M. The role of hepatic conversion from impaired glucose tolerance to type 2 diabetes mellitus:
lipase in lipoprotein metabolism and atherosclerosis. Curr Opin Lipidol the Finnish Diabetes Prevention Study. J Clin Endocrinol Metab
1998;9:211–9. 2004;89:2019–23.

Diet modifies LIPC gene effect on serum lipids 1293


19. Lindi V, Schwab U, Louheranta A, Vessby B, Hermansen K, Tapsell L, 23. Qi L. Mendelian randomization in nutritional epidemiology. Nutr Rev
Riccardi G, Rivellese AA, Laakso M, Uusitupa MI; KANWU Study 2009;67:439–50.
Group. The G-250A polymorphism in the hepatic lipase gene promoter 24. Voight BF, Peloso GM, Orho-Melander M, Frikke-Schmidt R, Barbalic
is associated with changes in hepatic lipase activity and LDL choles- M, Jensen MK, Hindy G, Hólm H, Ding EL, Johnson T, et al. Plasma
terol: the KANWU Study. Nutr Metab Cardiovasc Dis 2008;18:88–95. HDL cholesterol and risk of myocardial infarction: a Mendelian
20. Brunzell JD, Zambon A, Deeb SS. The effect of hepatic lipase on randomisation study. Lancet 2012;380:572–80.
coronary artery disease in humans is influenced by the underlying 25. Deeb SS, Peng R. The C-514T polymorphism in the human hepatic
lipoprotein phenotype. Biochim Biophys Acta 2012;1821:365–72. lipase gene promoter diminishes its activity. J Lipid Res 2000;41:
21. Cahova M, Dankova H, Palenickova E, Papackova Z, Kazdova L. The 155–8.
opposite effects of high-sucrose and high-fat diet on fatty acid oxidation 26. Tahvanainen E, Syvanne M, Frick MH, Murtomaki-Repo S, Antikainen
and very low density lipoprotein secretion in rat model of metabolic M, Kesaniemi YA, Kauma H, Pasternak A, Taskinen MR, Ehnholm C.
syndrome. J Nutr Metab 2012;2012:757205. Association of variation in hepatic lipase activity with promoter
22. Qi L, Cho YA. Gene-environment interaction and obesity. Nutr Rev variation in the hepatic lipase gene. The LOCAT Study Investigators.
2008;66:684–94. J Clin Invest 1998;101:956–60.

Downloaded from https://academic.oup.com/jn/article-abstract/145/6/1289/4616777 by guest on 13 June 2020

1294 Xu et al.

Das könnte Ihnen auch gefallen